Bootstrapping is a cheap way for academics to do proof-of-concept studies in MS #MSBlog #MSResearch
“Testing established drugs or treatment paradigms in diseases are difficult. This typically starts off with a clinician simply trying a drug in a person with a disease and seeing how they do. If they respond he/she would then treat a few more people and put together a case series. If the case series of responders was convincing it would form the basis for doing a proof-of-concept study. MS is full of examples of this approach, e.g. interferon-beta, rituximab (anti-CD20), alemtuzumab (anti-CD52) and daclizumab (anti-CD25) therapy. The following is the proof of concept study that led to the development programme for daclizumab.”
Identifying effective treatment combinations for MSers failing standard therapy is an important goal. This study reports the results of a phase II open label baseline-to-treatment trial of a humanized monoclonal antibody against CD25 (daclizumab) in 10 MSers with incomplete response to IFN-beta therapy and high brain inflammatory and clinical disease activity. Daclizumab was very well tolerated and led to a 78% reduction in new contrast-enhancing lesions and to a significant improvement in several clinical outcome measures.
“As you can see in this picture the original daclizumab trial was an open-label study, i.e. everyone gets treated with daclizumab and you simply count the number of gadolinium-enhancing lesions on MRI before and after treatment with daclizumab to assess whether or not the treatment is effective. At the end of the study the MRI images are assessed blind by a radiologist and the number of enhancing lesions are counted. This is called a bootstrap study. The MRI experts have now developed well defined protocols with power calculations, which allows us to use this type of study in early proof-of-concept studies. They are called bootstrap studies as MSers cross over from a period of observation to active treatment and they are not necessarily blinded. Bootstrap proof-of-concept studies are typical for academic or investigator-led studies because they are relatively cheap to perform (~$1.0-1.5M). Industry on the other hand prefers to do parallel double-blind placebo-controlled studies, which cost about 5-10x more. The table below is from the article that gives the numbers of MSers needed in these studies to get enough power to test whether or not a drug is effective or not.”
“Please note we are using the bootstrap design in our INSPIRE trial (raltegravir in MS), as part of the Charcot Project, to see whether or not raltegravir reduces MRI activity in MSers with active scans. We have powered the study to determine a 50% treatment effect in RRMSers with active baseline scans; this is why we are aiming to recruit 24 subjects. Please note we are currently testing raltegravir in an ethically approved study; all study subjects have to give informed consent and have their costs of participating covered. We do not advocate using raltegravir off-license to treat MS. If the INSPIRE trial is negative we will know that raltegravir does not have a major effect on MS disease activity. If on the other hand it is positive we will then have to discuss with Merck, the manufacturer of the drug, the next steps in terms of doing larger more definitive studies, i.e. a phase 2b and 3 development programme. In MS the latter studies take anything from 8 to 12 years to complete. Drug development is slow process; unfortunately, the regulations do not allow for anything faster to occur. Drug development is all about maximising the risks and benefits for people with MS; there are no speedy solutions. In the case of daclizumab this proof-of-concept study was published in 2004. The phase 3 study is currently running and I would expect that if positive and the drug is licensed it will be with us in late 2016 or early 2017; 12-13 years after the publication of the proof-of-concept study.”
OBJECTIVE: A new parametric simulation procedure based on the negative binomial (NB) model was used to evaluate the sample sizes needed to achieve optimal statistical powers for parallel groups (with (PGB) and without (PG) a baseline correction scan). It was also used for baseline versus treatment (BVT) design clinical trials in relapsing-remitting (RR) and secondary progressive (SP) multiple sclerosis (MS), when using the number of new enhancing lesions seen on monthly MRI of the brain as the measure of outcome.
METHODS: MRI data obtained from 120 untreated RRMSers selected for the presence of MRI activity at baseline, 66 untreated and unselected RRMSers, and 81 untreated and unselected SPMSers were fitted using an NB distribution. All these MSers were scanned monthly for at least 6 months and were all from the placebo arms of three large scale clinical trials and one natural history study. The statistical powers were calculated for durations of follow up of 3 and 6 months.
RESULTS: The frequency of new enhancing lesions in patients with SPMS was lower, but not significantly different, from that seen in unselected RRMSers. As expected, enhancement was more frequent in RRMSers selected for MRI activity at baseline than in the other two MSer groups. As a consequence, the estimated sample sizes needed to detect treatment efficacy in selected RRMSers were smaller than those of unselected RRMSers and those with SPMS. Baseline correction was also seen to reduce the sample sizes of PG design trials. An increased number of scans reduced the sample sizes needed to perform BVT trials, whereas the gain in power was less evident in PG and PGB trials.
CONCLUSION: This study provides reliable estimates of the sample sizes needed to perform MRI monitored clinical trials in the major MS clinical phenotypes, which should be useful for planning future studies.